Development of a continuous process for α-thio-β-chloroacrylamide synthesis with enhanced control of a cascade transformation

Article abstract of DOI:10.3762/bjoc.12.246

A continuous process strategy has been developed for the preparation of α-thio-β-chloroacrylamides, a class of highly versatile synthetic intermediates. Flow platforms to generate the α-chloroamide and α-thioamide precursors were successfully adopted, progressing from the previously employed batch chemistry, and in both instances afford a readily scalable methodology. The implementation of the key α-thio-β-chloroacrylamide casade as a continuous flow reaction on a multi-gram scale is described, while the tuneable nature of the cascade, facilitated by continuous processing, is highlighted by selective generation of established intermediates and byproducts.

Full text of DOI:10.3762/bjoc.12.246

Development of a continuous process for α-thio-β-
chloroacrylamide synthesis with enhanced
control of a cascade transformation
Olga C. Dennehy1, Valérie M. Y. Cacheux1, Benjamin J. Deadman1,§, Denis Lynch1,
Stuart G. Collins*1, Humphrey A. Moynihan*1 and Anita R. Maguire*2
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2016, 12, 2511–2522.
1Department of Chemistry, Analytical and Biological Chemistry
doi:10.3762/bjoc.12.246
Research Facility, Synthesis and Solid State Pharmaceutical Centre,
University College Cork, Cork, Ireland and 2Department of Chemistry
and School of Pharmacy, Analytical and Biological Chemistry
Research Facility, Synthesis and Solid State Pharmaceutical Centre,
University College Cork, Cork, Ireland
Received: 22 September 2016
Accepted: 04 November 2016
Published: 24 November 2016
This article is part of the Thematic Series "Automated chemical
synthesis".
Email:
Stuart G. Collins* - stuart.collins@ucc.ie; Humphrey A. Moynihan* -
H.Moynihan@ucc.ie; Anita R. Maguire* - a.maguire@ucc.ie
Guest Editor: I. R. Baxendale
*
Corresponding author
© 2016 Dennehy et al.; licensee Beilstein-Institut.
License and terms: see end of document.
§
Current address: Department of Chemistry, Imperial College
London, U.K.
Keywords:
α-thio-β-chloroacrylamides; cascade reactions; flow chemistry
Abstract
A continuous process strategy has been developed for the preparation of α-thio-β-chloroacrylamides, a class of highly versatile syn-
thetic intermediates. Flow platforms to generate the α-chloroamide and α-thioamide precursors were successfully adopted,
progressing from the previously employed batch chemistry, and in both instances afford a readily scalable methodology. The imple-
mentation of the key α-thio-β-chloroacrylamide casade as a continuous flow reaction on a multi-gram scale is described, while the
tuneable nature of the cascade, facilitated by continuous processing, is highlighted by selective generation of established intermedi-
ates and byproducts.
Introduction
Since the efficient and highly stereoselective transformation of combined influence of the amide and chloro substituents, miti-
α-thioamides to the corresponding α-thio-β-chloroacrylamides gating the electron-donating effect of the sulfide moiety.
derivatives was first reported [1,2], the considerable synthetic Nucleophilic substitution [4], Diels–Alder reactions [5] and 1,3-
utility of these heavily functionalized acrylamide compounds dipolar cycloadditions [6-9], and oxidation of the sulfide group
has been well documented [3]. The predominant site of reactivi- [10-12] are among a wide array of transformations which have
ty is at the electrophilic β-carbon, which results from the been successfully applied to these compounds (Scheme 1).
2511

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
matographic purification. Finally, the optimized conditions for
the final cascade transformation employ rapid heating via ‘hot-
plunge’ in order to minimize the formation of process impuri-
ties during the initial heating phase [1]. This efficient rapid
heating poses practical difficulties for scale-up and, further-
more, chromatographic separation is required to remove prod-
uct impurities.
The nature of the aforementioned difficulties outlined are, how-
ever, largely specific to the scale-up of batch chemistry. A con-
tinuous processing approach frequently possesses advantages
over the batch equivalent, as has been extensively documented
[
14-22]. When combined with automated operation, it allows
for enhanced reproducibility and access to extreme conditions,
which, along with improved heat and mass transfer, all facili-
tate significant ease of scale-up. The reaction control afforded
by use of high surface-area-to-volume ratio tubular reactors,
specifically with respect to dissipation of heat, offers a safety
profile unique to flow chemistry. Continuous processing also
provides the capacity to continuously generate hazardous
Scheme 1: Reaction pathways of α-thio-β-chloroacrylamides.
In order to fully exploit the synthetic potential of these reagents and intermediates in small quantities, in situ, and trans-
β-chloroacrylamides, however, a means of ready access to ferred directly into a reaction stage without operator handling
appreciable quantities of material is required. Preparation of [21-26]. As rapid heat transfer (steps 1 and 3) and greater reac-
α-thio-β-chloroacrylamides typically results from a three-step tion control (steps 2 and 3) were identified as the key chal-
synthetic route, culminating in a final cascade/domino reaction lenges to be overcome, we envisaged that continuous process-
[
13] where a toluene solution of α-thioamide and NCS is sub- ing could facilitate the preparation of large quantities of α-thio-
jected to a ‘hot plunge’ by placing it into an oil bath at 90 °C β-chloroacrylamide with reduced purification requirements. The
Scheme 2). While this route has consistently provided a robust goal of this study was to develop an optimized process for the
(
means of generating the desired β-chloroacrylamides at scales synthesis of α-thio-β-chloroacrylamides, employing a model
of 1–10 g, it suffers from several disadvantages which impact system with N-4′-methylphenyl-(Z)-3-chloro-2-(phenyl-
on the ease of scale-up.
thio)propenamide (Z-3) as the target product. This optimized
process would utilise flow chemistry as a key enabling technol-
The preparation of the α-chloroamide 1 is exothermic and ogy to overcome the aforementioned challenges.
requires significant external cooling, an undesirable feature for
Results and Discussion
scale-up. The synthesis of the α-thioamide 2 involves prior gen-
eration of fresh sodium ethoxide from sodium metal. Further- Preparation of α-chloroamide
more, this α-thioamide protocol, at high pH, ordinarily does not The synthesis of α-chloroamide 1 is highly exothermic, due to
go to completion, leaving unreacted starting material and the neutralisation of HCl – a byproduct – with triethylamine,
forming impurities which are subsequently removed by chro- and the need for effective heat removal imposes limitations on
Scheme 2: Typical three-step batch preparation of α-thio-β-chloroacrylamide.
2512

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
the batch scale-up of this step. It was envisaged that the effi-
cient heat transfer properties of a high surface area tubular flow
reactor would remove the need for external cooling of the reac-
tion. To facilitate safe scale-up of this reaction we initially in-
vestigated a direct transfer of the batch process (Scheme 3) to
continuous mode.
Scheme 3: Batch process for preparation of α-chloroamide 1.
Although initial investigations involving small throughput had
shown promise, the practicalities of employing dichloro-
methane at process scales caused us to consider alternative
‘
greener’ solvent systems [27]. A screen of alternative solvents
in batch test reactions revealed that, while the amide formation
was tolerant to most solvents, rapid precipitation of triethyl-
amine hydrochloride would be problematic in a continuous
process. Indeed, trial runs of a continuous process in ethyl
acetate resulted in immediate blockage of the flow reactor at the
Scheme 4: Process for the conversion of 2-chloropropionyl chloride
and p-toluidine to α-chloroamide 1 under optimized flow conditions.
point of reagent mixing. To prevent blockages due to salt for- ered. Attempts were made to use triethylamine as the base in a
mation we investigated replacements for triethylamine that continuous thiolation reaction, however, the reaction was found
would produce a more soluble HCl salt. Diisopropylethylamine to progress slowly and a maximum conversion of 39% was ob-
(
DIPEA) was found to be a suitable base that allowed the con- served by 1H NMR spectroscopy. Elevating the temperature to
tinuous process to be carried out in ethyl acetate without any 90 °C and employing DBU, as a more basic alternative to tri-
observed precipitation of the HCl salt. The ‘greener’ continu- ethylamine, did not increase the reactivity. Hence, the focus was
ous amide formation (Scheme 4) was carried out on a large instead directed on converting the existing batch process
scale, producing 91 g (92% yield) of the α-chloroamide 1 over (Scheme 2), with sodium ethoxide as base, into a stand-alone
5
hours of continuous operation, as a white crystalline solid continuous process.
after aqueous work-up and recrystallization.
Initially, however, the sodium chloride byproduct was found to
precipitate from ethanol causing blockages at the back-pressure
Synthesis of α-thioamide
Driving the reaction to completion and avoiding the use of sodi- regulator. As sodium chloride possesses a relatively low solu-
um metal were the key aims in transferring α-thioamide prepa- bility in ethanol (ca. 0.055 g in 100 g of ethanol at 20 °C) com-
ration from batch to flow. Although yields of 80–90% can be pared to methanol (1.375 g in 100 g) [28], methanol was pro-
obtained under batch conditions, incomplete conversion to posed as an alternative solvent. As 1H NMR analysis indicated
α-thioamide 2 necessitates a difficult, and often laborious, chro- that the crude reaction product from batch tests (using metha-
matographic separation, as starting material 1 and product 2 are nol as solvent) consisted of 98% α-thioamide 2, the process
poorly resolved. Indeed, high-purity batches of α-thioamide 2 was subsequently transferred to a continuous flow system
are often not achieved by chromatography, with the resulting (Scheme 5).
product typically ca. 94% pure by HPLC. It was also envisaged
that the facility to superheat the solvent in a pressurised contin- A variety of temperatures (60–120 °C), bases (NaOMe, NaOH,
uous platform could enable sodium ethoxide to be replaced by a Na2CO3) and concentrations (0.1–0.3 M) were investigated
weaker base, obviating the need for sodium metal.
using methanol as solvent (see Supporting Information File 1),
however, unreacted α-chloroamide 1 and diphenyl disulfide
At an early stage of process development, the possibility of tele- were detected as product components in all experiments. Direct
scoping the amide formation and thiolation steps was consid- sampling of the reaction mixture (system effluents) also showed
2513

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
Ultimately, a residence time of 5 min at 120 °C, using a 0.25 M
concentration of α-chloroacrylamide, was found to give an
acceptable quality of product 2, with no detectable quantities of
starting material 1 or diphenyl disulfide by HPLC analysis
(
entry 13, Table 1).
The optimized continuous process (Scheme 6) was then run on
a 5 g scale with no observed loss of yield or purity. The
α-thioamide 2, which crystallized directly from the output of the
flow process, was obtained in 71% yield and found to be >99%
pure by HPLC analysis, compared to 94% purity for a typical
batch preparation following chromatography.
Scheme 5: Conversion of 1 to 2 in continuous mode using MeOH as
solvent.
additional component peaks by HPLC analysis, which were not
observed in material isolated after the reaction work-up. While
temperatures above 100 °C or α-chloroamide 1 concentrations
above 0.1 M were not found to be advantageous, sodium
hydroxide demonstrated promising results when used as base.
With use of sodium hydroxide in mind, replacement of metha-
nol with an ethanol/water mixture as solvent was subsequently
examined. This solvent change was investigated in conjunction
with further refinements to the stoichiometry of sodium hydrox-
ide and thiophenol used, along with optimization of process
temperature and residence time (Table 1).
Scheme 6: Optimized process for the conversion of α-chloroamide 1
to α-thioamide 2 under flow conditions.
Initially, when using 10 equivalents of sodium hydroxide, the As transferring the α-thioamide preparation to a continuous
best conversion to product 2 was obtained at a reaction tempera- platform had involved a number of important changes to the
ture of 120 °C (entry 2, Table 1), with no unreacted reaction conditions, it was decided to evaluate the optimized
α-chloroamide 1 detected by HPLC. Employing just 5 equiva- flow conditions (entry 13, Table 1) when applied to a batch
lents of hydroxide also provided an acceptable yield of process for comparison: 1.05 equivalents of thiophenol in
α-thioamide 2 in all instances (entries 4–13, Table 1). The use ethanol mixed with an aqueous solution containing 5 equiva-
of an excess of sodium hydroxide as base had removed the lents of sodium hydroxide followed by heating to reflux for
difficulty with unreacted starting material, presumably by 1 hour. Initially on a 500 mg scale, a 97% yield of α-thioamide
hydrolysis of unreacted α-chloroamide 1 to more water soluble 2 was obtained, while operating at a higher concentration (in-
byproducts. In order to minimize the presence of diphenyl disul- creased from 0.25 M α-chloroamide 1 to 0.4 M α-chloroamide
fide in the isolated product, the stoichiometry of thiophenol was 1) a yield of 94% was achieved on a 5 g scale, with the isolated
also examined. Interestingly, a reduction in the excess of thio- product determined to be 99% pure by HPLC analysis. This
phenol to 1.05 equivalents was found to give a greater propor- process was ultimately carried out at a 20 g scale achieving
tion of α-thioamide 2 and significantly reduced level of 88% yield, again with 99% purity; the decrease in yield was
diphenyl disulfide (entries 7–13, Table 1).
offset by the increase in both productivity at this scale and prod-
uct purity. The ability to operate effectively at higher concentra-
After an improved stoichiometry of reagents had been estab- tions in batch than in flow, in this case, made this batch process
lished, lowering the residence time was investigated to facili- the optimum method of α-thioamide preparation, with a consid-
tate efficient large scale synthesis by a continuous flow process. erable reduction in reaction time from 10 hours to just 1 hour
2514

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
Table 1: Optimization of temperature, thiophenol concentration, residence time and stoichiometry of base for conversion of 1 to 2 in continuous
modea using EtOH/H2O as solvent.
Entry
Residence
time
Temp.
(°C)
PhSH
(equiv)
NaOH
(equiv)
Product ratio
2
%)b
1
PhSSPh
(%)b
Otherc
(%)b
(min)
(
(%)b
1
2
3
4
5
6
7
8
9
30
30
30
30
30
30
30
10
10
5
100
120
140
100
100
100
100
100
120
120
120
140
120
1.4
1.4
10
10
10
5
73.2
82.2
54.5
75.9
78.8
81.0
85.4
67.4
77.3
81.0
72.8
71.3
74.1
0
0
2.4
6.1
1.3
2.6
3.5
0.6
1.1
0.4
0.9
1.2
1.1
1.3
0
24.4
11.7
44.2
21.5
14.2
18.0
13.5
23.7
21.8
15.7
21.6
27.4
25.9
1.4
0
1.4
0
1.2
5
3.6
0.4
0
1.1
5
1.05
1.05
1.05
1.05
1.05
1.05
1.05
5
5
8.5
0
5
10
11
12
5
2.1
4.5
0
2
5
2
5
13d
5
5
0
aGeneral conditions: 1 equiv α-chloroamide 1 (2 mL of a 0.1 M solution in EtOH) was reacted with PhSH (as a solution in EtOH) and NaOH (as a solu-
tion in H2O). bDetermined by HPLC analysis (peak area: see Supporting Information File 1) of samples taken directly from flow reactor as effluent
solutions and diluted in MeCN prior to analysis. cUnisolated components, not present after work-up. dReaction was run using 2 mL 0.25 M solution of
α-chloroamide 1 in EtOH.
(
for 20 g of 2) and with a reduction of approximately one third The stoichiometry of sodium hydroxide required for reaction
in the required solvent volume, compared to the flow process. completion was also considered as part of the batch comparison.
By comparison, the original batch process was typically run for Here, a reduction from 5 equivalents to 3 and subsequently to
2
0 hours on scales up to 10 g.
just 2 equivalents was found to be possible, with no discernible
negative impact on the product formation. In the latter case, in
As with the optimized flow process, direct crystallisation of the batch the α-thioamide 2 was recovered in 92% yield and >99%
α-thioamide product 2 from this improved batch process was purity by HPLC, when the reaction was performed on a 5 g
achieved by cooling and adding water as anti-solvent. This scale. A subsequent batch run on a 20 g afforded an 89% yield,
method of product isolation obviated the need for the arduous with the same level of product purity. It is, perhaps, worth
work-up – involving extraction into dichloromethane and noting that the high isolated yields obtained from the scaled-up
several aqueous washes – associated with the original batch reactions strongly suggest that the substantial quantities of
version, and gave material which was 99% purity or greater by ‘other’ components observed by HPLC analysis, but removed
HPLC analysis.
during work-up, are overestimated by detection at 250 nm
2515

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
(
Table 1). Such an overestimation is consistent with the pres- the current process – in either batch or flow – indicated com-
ence of additional chromophores, when compared to the desired plete consumption of the α-chloroamide 1, without diphenyl
product, and would indicate that these observed components disulfide formation, while an increase in yield from 80–90% to
may contain an α,β-unsaturated carbonyl motif in their struc- consistently over 90% has been achieved. Furthermore, the use
tures.
of an inert atmosphere is no longer necessary as the decrease in
reaction time has essentially eliminated the opportunity for
The value of exploring flow methodology, ultimately leading to aerobic oxidation of the thiophenolate anion to diphenyl disul-
an improved batch process, is keenly highlighted in this fide, while sodium metal is no longer used as part of the
instance. The optimized batch process, developed through process.
examining the use of continuous processing, can produce 20 g
of pure material, with direct product precipitation/crystalliza- α-Thio-β-chloroacrylamide cascade in flow
tion from the reaction solution (>99% pure by HPLC analysis), Successful conversion of the β-chloroacrylamide cascade step
which has removed the requirements for isolation by extraction from batch to flow posed a number of challenges. The reaction
and subsequent chromatographic purification. HPLC analysis of mechanism (Scheme 7) involves a complex cascade which also
Scheme 7: Mechanism of the β-chloroacrylamide cascade process [29].
2516

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
gives rise to several known impurities, including acrylamide 4, nent of the reactant stream (from α-thioamide 2 preparation),
dichloride 5, trichloride 6 and dichloroacrylamide 7.
particularly at elevated temperatures, strongly mitigated against
integrating these steps.
In the optimized batch synthesis of α-thio-β-chloroacrylamide
Z-3 from the corresponding α-thioamide 2, N-chlorosuccin- For the batch process, the solubility of NCS in toluene has
imide (NCS) is added in one portion to a solution of 2 in tolu- notable benefits: NCS is soluble in toluene at high temperatures,
ene and the reaction mixture is immediately immersed in an oil while the succinimide byproduct readily precipitates from tolu-
bath at 90 °C (Scheme 2). Although this protocol performs well, ene on cooling, allowing its convenient removal by filtration. In
giving 91% yield on a ca. 5 g scale [1], the practical challenges a continuous flow process, however, succinimide precipitation
of achieving efficient rapid heating on a larger scale in batch would cause blockage of the system.
made continuous processing an attractive alternative for scale-
up due to its capacity for excellent temperature control. Effi- Attempts at transferring the cascade reaction to a continuous
cient heat transfer due to the high surface, low volume geome- platform began with direct adaptation of the existing batch
try of tubular flow reactors makes it possible to achieve process (Table 2). The solubility of NCS in toluene was found
extremely rapid temperature transitions. It was envisaged that to be variable and often unsuitably low. Only batches of NCS
flowing the reaction through a heated section of tubing would which readily gave complete solutions were used and these
be analogous to the batch ‘hot plunge’ method but with the batches were always either freshly recrystallized or commercial
capacity for faster heating of the reaction.
batches which were ‘newly’ opened prior to use. The reduced
solubility of other batches was attributed to the partial hydroly-
Given the superior performance of α-thioamide 2 synthesis in sis of NCS upon intermittent exposure to ambient conditions
batch, the potential telescoping of the thiolation process with over prolonged periods, also generating HCl.
the β-chloroacrylamide cascade was not investigated. Further-
more, the potential vulnerability of α-thio-β-chloroacrylamides Initial investigations using our prototype flow process em-
towards nucleophilic substitution by an aqueous ethanol compo- ployed 0.01 M solutions of NCS and starting material 2 in tolu-
Table 2: Initial flow process for conversion of 2 to Z-3 using toluene as solvent.
Entry
Ratio
Residence Time
(min)
Product ratio
2:NCSa
2
4
5
Z-3
(%)b
(%)b
(%)b
(%)b
1
2
3
4
5
1:2
1:2
20
50
20
20
20
19
21
22
25
21
15
19
8
18
0
47
60
46
62
2
1:2.3
1:3
20
12
<1
1
1:1
77
aStoichiometric ratio of α-thioamide 2:NCS controlled by manipulating the relative flow rates. bMolar ratio determined by HPLC analysis (peak area
weighted for relative response factors of each component: see Supporting Information File 1) of samples taken directly from flow reactor as effluent
solutions and diluted in MeCN prior to analysis.
2517

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
ene. The equivalent of a ‘hot-plunge’ method was achieved by in the cascade reaction using a continuous process, more effec-
passing the reaction solution through a coiled tube reactor at tively than in batch and with greater flexibility.
1
20 °C. The high surface area–volume ratio of tubular flow
Optimization of the cascade process using
reactors is ideal for such rapid temperature transitions. It was
noted that a relatively short residence time of only 20 min could flow chemistry
be used, with a longer time of 50 min offering only a modest In all the aforementioned cases (Table 2), around 20% of the
improvement on the reaction outcome (entries 1 and 2, Table 2). starting material was consistently found to be unreacted. The
Indeed, the conversion of starting material 2 to acrylamide 4 key limitation to overcome was proposed to be the low solu-
was found to be closely comparable, indicating almost identical bility of NCS in toluene, and the consequent limitations to
reaction progress, given the instability of dichloride 5, which reactor throughput. To offset this difficulty, the use of alterna-
easily converts to the final product Z-3.
tive solvents was investigated. Acetonitrile was considered as a
possible alternative solvent due to the high solubility of NCS it
Increasing the amounts of NCS used was found to lead to a offers. Hence, preliminary experiments were carried out in
better conversion of acrylamide 4 to dichloride 5 or α-thio-β- order to compare its performance to toluene (Table 3), with the
chloroacrylamide Z-3 (entries 3 and 4, Table 2), with only 1% α-thioamide 2:NCS ratio again adjusted by manipulating the
of acrylamide 4 left unreacted with three equivalents of NCS concentration of the reagent solutions. In these experiments, the
used. When only an exact stoichiometric ratio (1:1) of NCS was reaction conversions were determined using 1H NMR analysis
used, the reaction stopped after the first chlorination step, of the crude product material obtained, with characteristic
leading to a reaction mixture which contained acrylamide 4 as proton signals of the β-carbon of the starting material 2, inter-
the main product formed (entry 5, Table 2). This ability to halt mediates 4 and 5, and the desired product Z-3 being easily iden-
the cascade at the acrylamide intermediate 4 or push through to tifiable.
the α-thio-β-chloroacrylamide Z-3 highlights the enhanced
control of reaction stoichiometry afforded by a continuous plat- Using toluene as a solvent for both reagents (α-thioamide 2 and
form and offers the possibility to isolate selected intermediates NCS) leads to 10% unreacted acrylamide 4 (entry 1, Table 3).
Table 3: Solvent screen for conversion of 2 to Z-3 in continuous mode.
Entry
[α-Thioamide 2]
[NCS]
(mM)
Solvent
A/B
Product ratio
(mM)
2
4
5
Z-3
E-3
(%)a
(%)a
(%)a
(%)a
(%)a
1
2
3
4
5
25
25
50
50
Tol/Tol
0
0
0
0
0
9.9
4.3
0
9.9
0
78.7
81.4
86.9
83.5
87.8
1.5
14.3
13.1
9.2
Tol/MeCN
25
50
MeCN/MeCN
Tol/MeCN
0
200
200
400
400
7.3
0
0
MeCN/MeCN
0
12.2
aDetermined by 1H NMR spectroscopy.
2518

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
When a solution of 2 in toluene and a solution of NCS in aceto- complete absence of the over-chlorinated products 6 and 7,
nitrile were employed as the reactant streams, similar results which were not observed by HPLC analysis or 1H NMR spec-
were observed at either low or high concentration, in terms of troscopy. In contrast, when similar conditions were employed in
residual acrylamide intermediate detected (entries 2 and 4, batch, significant formation of these byproducts was often
Table 3). However when acetonitrile was used as solvent for in evidence [1]. The flow process for the conversion of
both reagents (α-thioamide 2 and NCS), full conversion to the α-thioamide 2 to α-thio-β-chloroacrylamide 3 which employed
final product Z-3 was observed, at both high and low concentra- acetonitrile as solvent was therefore taken forward for optimiza-
tion of reagents (entries 3 and 5, Table 3). Use of high concen- tion and scale-up (Table 4).
trations has the advantage of increasing process productivity. In
this case (entry 5, Table 3), the production could be increased The residence time of the flow process was investigated to de-
eight-fold for the same reaction time as entry 3 (25 min resi- termine the completion time of the reaction, principally to mini-
dence time). Furthermore, higher concentration of reagents mize the extent of impurity formation due to over-reaction. The
enables greener synthesis by reducing solvent use.
shortest possible effective residence time would be also prefer-
able for larger scale operation in order to maximize the reactor
Interestingly, during development studies on the conversion of throughput. The dichloride intermediate 5 was still present after
α-thioamide 2 to α-thio-β-chloroacrylamide Z-3 in acetonitrile, a 2 min residence time (entry 5, Table 4), implying the reaction
by flow or in batch, a new component of the cascade reaction had not yet reached completion, while minor impurities re-
was observed, which was identified as the (E)-α-thio-β- mained at similar levels throughout all of the experiments. The
chloroacrylamide E-3.
succinimide byproduct was removed in the product work-up.
An important feature of the experiments conducted on the At lower reaction temperatures, large quantities of the dichlo-
β-chloroacrylamide cascade as a continuous process was the ride 5 were observed (entries 6–8, Table 4), with correspond-
Table 4: Optimization of flow rates, residence time and temperature for conversion of 2 to Z-3 in continuous modea.
Entry
Residence
time
Flow rate
(mL/min)
Temp
(°C)
Product ratiob
E-3
Z-3
4
5
(%)c
(min)
(%)c
(%)c
(%)c
1
2
3
4
5
6
7
8
9
25
15
10
5
0.2
0.3
0.5
1.0
2.5
2.5
2.5
2.5
2.5
120
120
120
120
120
80
88.2
87.6
86.9
85.5
72.8
4.9
11.8
12.4
13.1
14.5
13.7
2.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2
13.5
92.7
87.4
78.3
0.0
2
2
90
9.5
3.1
2
100
130
17.3
84.0
4.4
2
16.0
a1 Equiv of α-thioamide 2 (4 mL of a 0.2 M solution in MeCN) was reacted with 2 equiv of NCS (4 mL of a 0.4 M solution in MeCN). bUnisolated com-
ponents, not present after work-up were not included, but ranged from 5–10% by peak area. cMolar ratio determined by HPLC analysis (peak area
weighted for relative response factors of each component: see Supporting Information File 1) of samples taken directly from flow reactor as effluent
solutions and diluted in MeCN prior to analysis.
2519

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
ingly low quantities of product Z-3. This finding is consistent consist only of a mixture of the Z- and E-isomers by 1H NMR
with previous work showing that rapid heating resulted in a spectroscopy, with pure Z-3 selectively recovered after recrys-
more efficient reaction cascade to the desired product Z-3, while tallization, albeit with a loss of isolated yield from this process.
slower heating leads to substantial quantities of reaction inter- This is the first instance in which multi-gram quantities of the
mediates 4 and 5 as product impurities [3].
product Z-3 have been isolated without the need for chromatog-
raphy and on more than 3 times the scale which can be ob-
The stoichiometry of NCS used for the continuous process was tained in batch with the same reaction time [1]; the increase in
also further optimized (Table 5). It was found that, at 130 °C, quantity and the ease of purification compensates for the reduc-
2
equivalents of NCS resulted in the lowest levels of the impuri- tion in yield to 57%. The material obtained by concentration of
ties arising from reaction intermediates and over-chlorination the liquors recovered from recrystallization were found to
byproducts while also achieving one of the highest conversions consist mainly of Z-3 and E-3 by 1H NMR spectroscopy. Purifi-
to the desired α-thio-β-chloroacrylamide Z-3 (82.9%, entry 4, cation of this material by chromatography gave an additional
Table 5).
11% yield of pure Z-3 (3.7 g).
This process was then operated on a 30 g scale (Scheme 8) to Conclusion
produce 19.3 g (57% yield, >99% pure by HPLC analysis and An efficient continuous flow methodology has been developed
1
H NMR spectroscopy) of isolated α-thio-β-chloroacrylamide for the three-step synthesis of α-thio-β-chloroacrylamide Z-3,
Z-3 in less than 4 hours. The crude material was found to which has overcome the challenges to scale-up posed by the
Table 5: Optimization of NCS stoichiometry for conversion of α-thioamide 2 to α-thio-β-chloroacrylamide Z-3 in continuous modea.
Entry
NCS
Product ratiob
equiv
Z-3
E-3
4
5
6
(%)c
(%)c
(%)c
(%)c
(%)c
1
2
3
4
5
6
7
1.7
1.9
1.95
2
68.8
73.8
76.5
82.9
81.6
81.3
69.4
13.1
14.1
14.8
15.8
15.9
15.3
11.9
18.1
12.1
8.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.1
1.7
18.7
1.3
2.05
2.1
2.2
1.4
1.7
0.0
a1 Equiv of α-thioamide 2 (4 mL of 0.2 M solution in MeCN) was reacted with NCS (4 mL of solution in MeCN) at 130 °C for 2 min, using a flow rate of
.5 mL/min. bMolar ratio determined by HPLC analysis (peak area weighted for relative response factors of each component: see Supporting Informa-
2
tion File 1) of samples taken directly from flow reactor as effluent solutions and diluted in MeCN prior to analysis. cUnisolated components, not
present after work-up were not included, but ranged from 2–14% by peak area.
2520

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
ceutical, for the kind loan of one of the flow chemistry reactors
used in this work.
References
1
.
.
Murphy, M.; Lynch, D.; Schaeffer, M.; Kissane, M.; Chopra, J.;
O’Brien, E.; Ford, A.; Ferguson, G.; Maguire, A. R. Org. Biomol. Chem.
2007, 5, 1228–1241. doi:10.1039/B618540A
2
Maguire, A. R.; Murphy, M. E.; Schaeffer, M.; Ferguson, G.
Tetrahedron Lett. 1995, 36, 467–470.
doi:10.1016/0040-4039(94)02288-M
3
.
.
Kissane, M.; Maguire, A. R. Synlett 2011, 1212–1232.
doi:10.1055/s-0030-1260559
4
Kissane, M.; Murphy, M.; O’Brien, E.; Chopra, J.; Murphy, L.;
Collins, S. G.; Lawrence, S. E.; Maguire, A. R. Org. Biomol. Chem.
2011, 9, 2452–2472. doi:10.1039/c0ob00805b
5
6
7
8
9
1
.
.
.
.
.
Kissane, M.; Lynch, D.; Chopra, J.; Lawrence, S. E.; Maguire, A. R.
Org. Biomol. Chem. 2010, 8, 5602–5613. doi:10.1039/c0ob00368a
Kissane, M.; Murphy, M.; Lynch, D.; Ford, A.; Maguire, A. R.
Tetrahedron 2008, 64, 7639–7649. doi:10.1016/j.tet.2008.05.026
Kissane, M.; Lawrence, S. E.; Maguire, A. R. Org. Biomol. Chem.
2010, 8, 2735–2748. doi:10.1039/c002479a
Kissane, M.; Maguire, A. R. Chem. Soc. Rev. 2010, 39, 845–883.
doi:10.1039/B909358N
Scheme 8: Optimized flow process for conversion of α-thioamide 2 to
α-thio-β-chloroacrylamide Z-3.
Kissane, M.; Lawrence, S. E.; Maguire, A. R. Tetrahedron 2010, 66,
4564–4572. doi:10.1016/j.tet.2010.04.057
0.Kissane, M.; Lynch, D.; Chopra, J.; Lawrence, S. E.; Maguire, A. R.
Tetrahedron: Asymmetry 2008, 19, 1256–1273.
conventional batch preparation. This approach has yielded
improvements in process safety, significantly reduced reaction
times and increased product purity, obviating the need for chro-
matography. One process, preparation of α-thioamide 2 ulti-
mately proved most efficient in batch, though the investigations
performed in flow were critical to achieving the optimization.
The easy access to synthetically useful amounts, afforded by a
transfer to continuous processing, is expected to significantly
increase the attractiveness of harnessing the enormous potential
utility of α-thio-β-chloroacrylamides on a more widespread
basis. Perhaps the most powerful outcome is the ability to
control the β-chloroacrylamide cascade through continuous pro-
cessing, leading to selective recovery of individual components
of the reaction.
doi:10.1016/j.tetasy.2008.04.033
11.Kissane, M.; Lawrence, S. E.; Maguire, A. R. Tetrahedron: Asymmetry
2010, 21, 871–884. doi:10.1016/j.tetasy.2010.05.004
12.Kissane, M.; Murphy, M.; Lawrence, S. E.; Maguire, A. R.
Tetrahedron: Asymmetry 2010, 21, 2550–2558.
doi:10.1016/j.tetasy.2010.10.007
1
3.Tietze, L. F. Chem. Rev. 1996, 96, 115–136. doi:10.1021/cr950027e
4.Wegner, J.; Ceylan, S.; Kirschning, A. Adv. Synth. Catal. 2012, 354,
1
17–57. doi:10.1002/adsc.201100584
15.Wirth, T., Ed. Microreactors in Organic Chemistry and Catalysis;
Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2013.
1
1
1
1
2
6.Baxendale, I. R.; Brocken, L.; Mallia, C. J. Green Process. Synth.
013, 2, 211–230. doi:10.1515/gps-2013-0029
7.Pastre, J. C.; Browne, D. L.; Ley, S. V. Chem. Soc. Rev. 2013, 42,
849–8869. doi:10.1039/c3cs60246j
2
8
8.Webb, D.; Jamison, T. F. Chem. Sci. 2010, 1, 675–680.
doi:10.1039/c0sc00381f
Supporting Information
9.McQuade, D. T.; Seeberger, P. H. J. Org. Chem. 2013, 78, 6384–6389.
doi:10.1021/jo400583m
Supporting Information File 1
0.Hartman, R. L.; McMullen, J. P.; Jensen, K. F. Angew. Chem., Int. Ed.
General information, experimental procedures, analytical
data and copies of NMR spectra of all compounds.
2
011, 50, 7502–7519. doi:10.1002/anie.201004637
21.Gutmann, B.; Cantillo, D.; Kappe, C. O. Angew. Chem., Int. Ed. 2015,
4, 6688–6728. doi:10.1002/anie.201409318
5
[http://www.beilstein-journals.org/bjoc/content/
2
2.Movsisyan, M.; Delbeke, E. I. P.; Berton, J. K. E. T.; Battilocchio, C.;
Ley, S. V.; Stevens, C. V. Chem. Soc. Rev. 2016, 45, 4892–4928.
doi:10.1039/C5CS00902B
supplementary/1860-5397-12-246-S1.pdf]
23.Deadman, B. J.; Collins, S. G.; Maguire, A. R. Chem. – Eur. J. 2015,
Acknowledgements
21, 2298–2308. doi:10.1002/chem.201404348
This work was undertaken as part of the Synthesis and Solid
State Pharmaceutical Centre supported by the Science Founda-
tion Ireland (grant: SFI SSPC2 12/RC/2275). The authors
would like acknowledge the contribution of Janssen Pharma-
24.Müller, S. T. R.; Wirth, T. ChemSusChem 2015, 8, 245–250.
doi:10.1002/cssc.201402874
2521

Beilstein J. Org. Chem. 2016, 12, 2511–2522.
2
2
2
5.Deadman, B. J.; O’Mahony, R. M.; Lynch, D.; Crowley, D. C.;
Collins, S. G.; Maguire, A. R. Org. Biomol. Chem. 2016, 14,
3423–3431. doi:10.1039/C6OB00246C
6.Kopach, M. E.; Roberts, D. J.; Johnson, M. D.; McClary Groh, J.;
Adler, J. J.; Schafer, J. P.; Kobierski, M. E.; Trankle, W. G.
Green Chem. 2012, 14, 1524–1536. doi:10.1039/c2gc35050e
7.Prat, D.; Wells, A.; Hayler, J.; Sneddon, H.; McElroy, C. R.;
Abou-Shehada, S.; Dunn, P. J. Green Chem. 2016, 18, 288–296.
doi:10.1039/C5GC01008J
2
8.Pinho, S. P.; Macedo, E. A. J. Chem. Eng. Data 2005, 50, 29–32.
doi:10.1021/je049922y
2
9.Foley, D. A.; Doecke, C. W.; Buser, J. Y.; Merritt, J. M.; Murphy, L.;
Kissane, M.; Collins, S. G.; Maguire, A. R.; Kaerner, A. J. Org. Chem.
2011, 76, 9630–9640. doi:10.1021/jo201212p
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.12.246
2522